Acta Pædiatrica ISSN 0803–5253
REGULAR ARTICLE
Weight gain in childhood and blood lipids in adolescence Bernardo L Horta (
[email protected])1 , Cesar G Victora1 , Rosangela C. Lima2 , Paulo Post1 ˆ 1.Post-Graduate Programme in Epidemiology, Universidade Federal de Pelotas, Pelotas, RS, Brazil ´ 2.Universidade Catolica de Pelotas, Pelotas, RS, Brazil
Keywords Blood lipids, Catch-up growth, Cholesterol, Cohort
Abstract Aim: To assess the effect of weight gain in childhood on blood lipid levels in adolescence.
Correspondence Bernardo L. Horta, Rua Triunfo 3080, 96090-790, Pelotas, RS, Brazil. Tel: +55 53 32841300 | Fax: +55 53 32841300 | Email:
[email protected]
Methods: A population-based birth cohort carried out in Pelotas, Southern Brazil. All newborns in the city’s hospitals were enrolled in 1982. The subjects have been followed up for several times in childhood. At age 18, 79% of all males were followed, and 2083 blood samples were available.
Received 24 June 2008; revised 24 November 2008; accepted 27 January 2009. DOI:10.1111/j.1651-2227.2009.01247.x
Re-use of this article is permitted in accordance with the Creative Commons Deed, Attribution 2.5, which does not permit commercial exploitation.
Adjusted analyses controlled for household assets index, family income, parental schooling at birth, maternal smoking during pregnancy and breastfeeding duration. Results: Birth weight for gestational age and weight gain in the first 20 months was not associated with blood lipid levels in adolescence. On the other hand, those subjects whose weight gain from 20 to 42 months of age was faster than that predicted from birth weight and weight-for-age z-score at the mean age of 20 months had lower high-density lipoprotein cholesterol (HDL) cholesterol [−0.78 (95% confidence interval: −1.28; −0.29)] and higher very low-density lipoprotein cholesterol (VLDL) and low-density lipoprotein cholesterol (LDL)/HDL ratio in adolescence. After controlling for current body mass index (BMI), the regression coefficient for HDL cholesterol decreased from −0.78 mg/dL to −0.29 mg/dL (95% confidence interval: −1.00 to 0.05). Conclusion: Weight gain from 2 to 4 years is related to an atherogenic lipid profile in adolescence and this association is mediated by current BMI.
INTRODUCTION Intrauterine malnutrition and early life growth patterns may result in metabolic and physiological programming with lifelong effects on the risk of selected diseases, including several cardiovascular conditions (1). Raised blood concentration of cholesterol would be a potential mechanism underlying these associations. Huxley et al. (2) reviewed the evidence linking birth weight and blood lipid levels, concluding that impaired foetal growth had little effect. In another meta-analysis, Lauren et al. (3) reached similar conclusions, but noted that after adjustment for current size, low birth weight became strongly related to total cholesterol, apparently supporting a programming effect of birth weight. However, as pointed by Lucas et al. (4) birth size adjusted for current size is a measure of change in size (centile crossing) between these two ages. Therefore, the observation by Lauren et al. (3) is compatible with a more marked effect of postnatal than of intrauterine growth. Weight gain up to adulthood has been positively associated with total cholesterol or triglycerides or negatively with high-density lipoprotein cholesterol (HDL) cholesterol (5,6,7). On the other hand, Kajantie et al. (8) reported that slow weight gain during infancy was associated with lower HDL cholesterol and higher non-HDL cholesterol at age 55–70 years. With respect to height gain, Eriksson et al. (9) observed that rapid height gain from 7 to 15 years was
1024
related to reduced HDL cholesterol concentration in elderly individuals; whereas Miura et al. (10) observed that total cholesterol at 20 years was inversely correlated with height gain from age 3 to 20 years. In the same way, Skidmore et al. (11) reported that total and low-density lipoprotein cholesterol (LDL) cholesterol at 53 years were inversely related with height at 2 years and height velocity between 15 years and adulthood. Studies in upper-income countries have observed that small for gestational age (SGA) infants often show rapid growth in the first two years (12). In low and middle-income countries, rapid growth in early childhood may have clear short-term benefits. Victora et al. (13) reported that infants who were born SGA and presented accelerated growth in the first 20 months of life had lower mortality and fewer hospital admissions subsequently. On the other hand, several studies suggest that rapid growth later in childhood might increase the risk of coronary heart disease, hypertension (14) and insulin resistance (15). In view of the conflicting evidence, more studies are needed to assess the pros and cons of rapid growth in infancy, to help quantify the socalled catch-up dilemma (16). In particular, studies should investigate whether there are critical periods in infancy and childhood in which rapid growth is more strongly associated with negative outcomes in adulthood (17). This study was aimed at assessing the effect of rapid growth in different age ranges on blood lipid levels of men belonging to a birth cohort.
C 2009 The Author(s)/Journal Compilation C 2009 Foundation Acta Pædiatrica/Acta Pædiatrica 2009 98, pp. 1024–1028
Horta et al.
METHODS The study was carried out in Pelotas (current urban population 320 000) in Southern Brazil. The population is mostly white, of Southern European descent. Like Brazil as a whole, there are wide social inequalities in health in this population. The infant mortality rate in the birth cohort was 38 per thousand live births and at the age of 18 years, 17.4% of all male subjects were overweight (body mass index [BMI] above 25 kg/m2 ). All 5914 infants born alive in three maternity hospitals in 1982 (over 99% of all births in the city) were recruited and have been followed up on several occasions (18). There were 3037 males and 2877 females. Their mothers were interviewed on socio-economic, demographic and health-related variables. Birth weight was obtained by the hospital staff using paediatric scales that were calibrated weekly by the research team and low birth weight was defined as less than 2500 g. Gestational age was estimated from the mothers’ recall of their date of last menstrual period, and those subjects whose gestational age was 500
72.7 80.1 84.0 79.3 76.6
Maternal schooling at delivery (years) ≤4 5—8 9—11 ≥12
76.2 81.7 75.8 78.9
Birth weight in grams
